Irrational science and the virus of bureaucracy
How the better-safe-than-sorry principle will make us more sorry than safe
Febrile reporting of the arrival of a Covid-19 Second Wave in the UK at the beginning of Autumn came as no surprise since the mainstream media were taking relish in predicting it before the first wave was over. Accompanying the media fanfare is a battle between two groups of scientists over the government’s policy for managing the crisis. One group believes that a radical shift in government policy is required to deal with the virus. This group sees the current efforts to suppress the virus as both futile and damaging in their disproportionate effect on society as a whole, taking into account the known risk profile of the disease. Let’s call this group the Radicals. The other group is broadly content with the government’s current strategy of continued suppression of the virus and wishes to see restrictions kept in place or tightened if necessary. Let’s call this group the Suppressers. Both the Radicals and the Suppressers wrote open letters to the government’s Chief Medical Officers setting out their arguments.
A close examination of the letters reveals something disturbing: one group has departed from rational scientific reasoning. Equally disturbing is a media bias which clearly favours this group while painting the other group as eccentric crackpots. How and why should we judge the experts? Digging into this question exposes an unhealthy relationship between scientific irrationality and the ever-growing monster of bureaucracy that feeds on modern-day capitalism.
How shall we judge the experts?
Most of us lack qualifications in a field relevant to understanding the science of Covid such as epidemiology, virology or health care. But we owe it to ourselves to weigh the experts’ arguments on merit given the overwhelming influence they have on our lives at present. History has demonstrated that a tyranny of science over humanity is no better than a tyranny of priests. Too often, ‘science’ and ‘scientists’ are not the same thing and it is a mistake to assume that the application of science by scientists will always produce good outcomes. Covid continues to divide scientists on many core aspects — the treatment protocol, herd immunity, how much weight should be given to case statistics in measuring the threat to public health, the efficacy of masks, and how to extricate ourselves from the quicksand of ever-shifting quarantine corridors and diktat on how many people can congregate in a variety of situations.
Logic belongs to all of us: we owe it to ourselves to exercise it
The truth is you don’t need a science qualification to make a judgement on which group of scientists has presented the more clear-headed arguments. The letter from each group is not actually a clash of scientific theory but rather a set of strategic arguments for either a change in policy, in the case of the Radicals, or a continuation of restrictions, in the case of the Suppressers. What is needed to judge the merits of the arguments is a reasonable nose for detecting errors in logic. As a former auditor trained to find errors, I examined the arguments to see which set had the most holes in it. The results were surprising because I did not expect to see such a wide gap between two groups of scientists with similarly large amounts of grey matter.
The Radicals are led by three scientists and one economist — Professors Sunetra Gupta and Carl Heneghan from Oxford University, Prof Karol Sikora from University of Buckingham and Sam Williams of the consultancy Economic Insight. The 28 signatories to the Radicals’ letter are drawn from a variety of specialisms including epidemiology, medical health, economics, research and statistics.
The Suppressers’ letter is authored by Trisha Greenhalgh of Oxford University and has 21 or 22 signatories[i].
Radicals vs Suppressers: Examining the arguments in detail
The Radicals’ argument is that the government’s existing policy does not fit the known Covid risk profile and that it is a blunt instrument harming all age groups and sections of society, including those who are least at risk. They also state that the objective of current policy seems to be to keep suppressing virus transmission until a vaccine is available and that this policy is not feasible. They acknowledge that this is an assumption but view it as a reasonable one in the absence of a clearer indication from the government of what the objective of current policy is. They argue that a functioning society and economy will be one that learns to live with the virus rather than one that constantly struggles to suppress it. There are eight pillars to their argument:
- They criticise the current policy for its lack of clearly stated objectives. They acknowledge that at the start of lockdown there was a clear objective — flatten the curve and prevent the NHS from being overwhelmed — but believe this is no longer the case and that there does not appear to be a clearly stated aim, despite new restrictions being announced almost weekly. The absence of clearly stated objectives leads to a two-part failure: the government is not aiming to achieve any concrete goals, and it is impossible to measure performance against unstated objectives.
- Objectives must go beyond simplistic targets aimed solely at reducing Covid deaths since there are complex trade-offs to be made both within the healthcare system and across different sections of society and the economy. There has to be a cost-benefit analysis that is framed more broadly than Covid itself.
- Policy formulation should be evidence-based. In the absence of sufficient existing empirical evidence owing to the novel nature of the virus, emphasis should be placed on performance measurement which tracks what is actually occurring on the ground. Whatever metrics are used should be capable of being accurately measured and reported. Heavy reliance on the current metrics that seem to be driving policy — namely cases and R factor — is also problematic as these metrics are “subject to significant measurement and interpretation challenges”.
- It makes no sense to target the entire population with restrictive measures when we know that 89% of Covid mortalities occur in the over-65 age group and that 95% of deaths occur in those with pre-existing co-morbidities. The blunt-instrument approach is creating costs in those demographics least affected and these costs could be avoided with a more targeted approach.
- Some examples are provided of the large costs borne by society in general and specific groups as a result of a blanket policy applied to the whole population: on cancer alone there is an estimated 2 million delayed or missed cancer screenings and treatments which could translate into a loss of 60,000 lives; meanwhile, unemployment is projected to reach 11.9% by the end of the year, with public sector net debt rising to £2 trillion for the first time.
- Despite the enormous costs of these policies, their effectiveness in reducing Covid deaths is not clear and there is no observable correlation between policy measures taken to date and the profile of Covid deaths. The available evidence shows that lockdown occurred after the infection peak and, as such, deaths would not have continued rising without the lockdown.
- Targeted measures should therefore be used to shield the most vulnerable sections of the population. The Radicals point out that Germany had a similar case fatality rate in the elderly to other EU countries but was successful in limiting the total number of cases in the elderly as evidenced by its achievement of the lowest number of deaths per capita in the EU.
- The final point the Radicals make focuses on the government’s media campaign to alter behaviour, which they say may have had negative impacts on mental and physical health more generally. The government’s deliberately crafted messages to heighten perceptions of personal threat are counterproductive to public welfare. An objective communication of the risks faced by age group and demographic would be preferable.
The letter from Professor Trisha Greenhalgh and colleagues — the Suppressers — has seven points. My critique of each, with reference also to the counterarguments of the Radicals outlined above, is presented in italics:
1. The first point expresses strong support for continued suppression of the virus across the entire population, rejecting segmentation and shielding of the vulnerable. They offer four reasons for this:
a) They acknowledge the demographic risk profile of Covid but state that deaths have occurred in all age groups and in people with no pre-existing medical conditions. They cite the risk of Long Covid — the debilitating effects of Covid in some sufferers that persist long after the initial infection has passed — adding that it affects “tens of thousands of people in the UK”.
The existence of a markedly variable risk profile by age is a self-evident reason for directing resources in proportion to the risk borne by each demographic. It would be absurd if the NHS started screening all under-60s for dementia on the grounds that we are aware of rare cases of dementia occurring in mid-life and yet Greenhalgh et al see no contradictions in the continuation of a blanket policy purely on the grounds that some deaths have occurred in those less at risk. To ignore the known 89% death rate in over-65s and the 95% death rate in the co-morbidity group is, in risk management terms, obtuse.
Long Covid is an emerging phenomenon and there is currently insufficient evidence about its prevalence, severity and risk profile to formulate policy. The government announced on 5th July 2020 that it will launch the first study in the UK into the long-term effects of the illness. The Suppresers are advocating the continuation of a policy that is now known to be causing damage to wider society and the economy in order to address a risk that has not yet been quantified. An assessment of the proportionality of the current response to the magnitude of the Long Covid risk has not been conducted.
b) The group claims a targeted approach can’t work because “society is an open system” and that it would be “practically impossible” to cut a cohort off from the rest of society. They also cite problems with multi-generational households and grandparents looking after children sent home from school.
This argument is problematic because, firstly, we do have an elderly cohort that is, in a very real sense, already cut off from the rest of society by virtue of being in care homes. We also know that, in health care terms, this boundary is not impregnable but you cannot simply rule out the possibility of designing a set of measures that might increase security against the virus in care homes. Such a policy should aim to strike a balance between protecting the vulnerable from infection and reducing the health harms caused by severe isolation. Some middle ground has to be found that involves the elderly making a choice for themselves about their health and quality of life rather than being treated as prisoners with no options. With regard to multi-generational households, the starting point of the debate is whether Covid is such a threat to the fabric of society that these households should not be allowed to choose for themselves what measures they will take to address their risks, with government playing a supportive rather than a prescriptive role. Do we still live in a liberal democracy and, if so, can we not ask whether the blanket restrictions imposed are proportionate to the risks faced?
c) They claim that herd immunity is based on an unproven assumption that reinfection will not occur and that it is not known whether herd immunity will wane over months or years.
Again, this approach implies a zero risk tolerance, accompanied by the costs required to maintain zero risk, while we wait for an indeterminate period of time for the answer. Put differently, the probability and impact of a worst case scenario is not known, but the group advocates implementing measures and incurring costs to prevent this undefined worst case scenario. This argument also ignores the emerging scientific evidence about existing T-cell immunity. Crucially, it also ignores the fact that if we can never gain any level of natural immunity, then a vaccine will also be of little help since vaccines are predicated on generating immunity through a durable antibody or T cell response. They are effectively planning for a scenario in which herd immunity is not possible and, by implication, a scenario in which a vaccine will also be ineffective. This approach raises the spectre of a society hiding under its bed indefinitely. If we agree that such an approach is not feasible then the sooner we start planning and implementing measures to live with the virus, the better. Sweden appears to have achieved this.
d) They claim there are no examples of a “segmentation and shielding” policy having worked in any country.
Possibly true. But only because none have yet been developed and tried. The false logic implied here is that if no existing solution currently exists, no solution can ever be developed. If the absence of a solution to a problem was an effective barrier to progress, we’d still be swinging from branches in trees. By contrast, the Radicals are, not unreasonably, asking the government to apply their minds to solving a public health problem in a different way. The Suppressers are saying, “Don’t bother”.
2. They express sympathy with the desire for a return to normality but caution that normality will regrettably have to be compromised because “the pandemic is following complex system dynamics”. They go on to express a desire to move flexibly between “90% normality and 60% normality”.
Neither of the points made represents an inherent reason to suppress the virus with restrictions that disproportionately affect the entire population. If anything, they are merely statements of the blindingly obvious — namely that, if we opt for a continuation of tight restrictions, normality will be elusive. Looking closer at the points, the reference to “complex system dynamics” without an explanation to make this tangible is merely obfuscation using jargon. It does nothing to advance an understanding of why normality is unachievable in the same way that replying to the question, “Are you in a relationship?” with “It’s complicated” does nothing to elucidate the nature of the relationship or its complexity. The insertion of a subjective normality range is also unscientific. The achievement of “90% normality”, or possibly even 60%, is something we might all be able to agree on, if only we had an objective method to agree what 90% means. How this range will be ‘flexibly’ navigated is also anyone’s guess.
3. They state that controlling the virus will catalyse economic recovery.
This statement could either be false or true depending on the paradigm used for the definition of “controlling the virus”. Since the authors are advocating continued suppression, it is fair to assume their interpretation of “controlling the virus” implies tight restrictions, including possible return to lockdown. Notwithstanding that there is no evidence of a correlation between severity of restrictions and impact on Covid deaths, we know that tight restrictions including lockdowns have decimated economies and will continue to do so if used again. Irrefutable evidence of this is provided not only by a stand-alone examination of the UK situation but also by comparison of the UK to Sweden, which employed the least restrictive measures in the EU and has so far achieved the most favourable economic outcome. Under the author’s paradigm of controlling the virus, the statement is false. Controlling the virus has damaged and will continue to damage the economy.
If you use the Swedish paradigm of controlling the virus, which has also achieved a better health outcome than the UK in deaths per capita, then the statement could be true. Controlling the virus by learning to live with it will, all things being equal, catalyse economic recovery.
4. While “evidence accumulates for airborne transmission” of the virus, the Suppressers argue for continued use of “measures which would help control the virus while also promoting economic recovery”. These measures are listed as: mandating face coverings in indoor settings; improving ventilation in schools and workplaces; social distancing, and; discouraging large indoor gatherings.
Continued social distancing and restrictions on large indoor gatherings are measures which would suppress economic activity in nearly all of the entertainment and related downstream sectors of the economy, including many retail sectors. The claim that these measures would also be “promoting economic recovery” is therefore false. The goal in a rational society would be to balance the efficacy of measures to reduce virus spread against the cost of those measures, questionable efficacy combined with high cost being the worst of both worlds. The graph below is a comparison of the UK’s death rates per million to Sweden’s, which provides some indication of efficacy and cost since Sweden continues to apply a much lighter level of societal restriction while achieving lover levels of deaths per million. The graphical comparison to Sweden not only demonstrates the ineffectiveness of the UK’s much tighter restrictions but also demonstrates that the comparison to Sweden is valid since it shows that the virus spread followed an almost identical trajectory in both countries.
5. The Suppressers advocate restoring routine medical appointments as “we move beyond the acute phase of the pandemic”.
This is another statement of the blindingly obvious since no-one in their right mind would argue this point and it adds nothing to the debate about whether there is a need for a change in policy. In light of the point made by The Radicals about the potentially devasting impact of missed cancer screenings and in light of the estimated backlog of 10 million NHS referrals, to contemplate anything other than the restoration of routine medical appointments would be criminal.
6. They state that in a complex system we should not expect to see simple linear relationships between a policy intervention and a desired outcome. Rather, they say, several policy measures may collectively contribute to controlling the virus and that this would require “complex analytical tools and rich case explanations to elucidate”.
This is an attempt to lower expectations on how to measure the effects of policy interventions and thus to put up a barrier to measuring the government’s performance. This is unacceptable, regardless of the difficulties involved in measuring the effectiveness of policy interventions. I imagine that the proportion of the tax-paying public receptive to “rich case explanations” to assess the success of government policy is miniscule. In reality, it shouldn’t be that hard to measure effects when we are dealing with an infectious disease. Two concrete metrics not requiring “rich case explanations” and which the taxpayer would not dispute are hospital admissions and death rates. These would need to be balanced against metrics that indicate how well we are functioning as a society — indirect deaths caused by restrictions, unemployment, depression and mental health, backlogs in referrals and untreated illnesses to name a few. Again, a semblance of “normal” might look something like this first-hand report from an A&E doctor (Sebastian Rushworth) in Stockholm:
“When I sit in the tube on the way to and from work, it is packed with people. Maybe one in a hundred people is choosing to wear a face mask in public. In Stockholm, life is largely back to normal. If you look at the front pages of the tabloids, on many days there isn’t a single mention of Covid anywhere. As I write this (19th September 2020) the front pages of the two main tabloids have big spreads about arthritis and pensions. Apparently arthritis and pensions are currently more exciting than covid-19 in Sweden.”
7. By now it should be apparent that the Greenhalgh et al letter is a gradual descent into the depths of academic bloviation. The very worst of it was saved for last. So that there can be no doubt about my italicised critique which follows, here is their final argument in full:
“While it is always helpful to have more data and more evidence, we caution that in this complex and fast-moving pandemic, certainty is likely to remain elusive. “Facts” will be differently valued and differently interpreted by different experts and different interest groups. A research finding that is declared “best evidence” or “robust evidence” by one expert will be considered marginal or flawed by another expert. It is more important than ever to consider multiple perspectives on the issues and encourage interdisciplinary debate and peer review. While government must continue to support research, some decisions — as you will be well aware — will need to be made pragmatically in the face of uncertainty”.
This is a series of statements the effect of which is to create the impression of a foggy future in which certainty and “facts”, placed contemptuously in quote marks by the author, will be elusive and research findings highly contested and therefore of little value. Aside from the unscientific nature of a subjective prediction of the future landscape, this characterisation of an increasingly foggy future is baseless since the facts are likely to become clearer with the accumulation of data over time, not murkier.
In any event, what is their recommendation for meeting these challenges? They are to be met with meaningless academic bureaucratism in the form of “consider[ing] multiple perspectives” and “encourage[ing] interdisciplinary debate”.
And the winner is…
The thrust of the Radicals’ argument is that the blunt instrument approach is incongruent with the known risk profile of the disease and that such an approach is burdening the majority of society with costs that could be avoided by a more targeted approach. A targeted risk-based approach is in fact nothing more than an embrace of the already established and scientifically evidenced approach to dealing with a pandemic of this nature. As the likelihood of a second nationwide lockdown grows, it is crucial to understand that there is no scientific basis for either lockdowns or an approach that fails to target those most at risk. As Martin Kulldorff, Harvard epidemiologist and signatory to the Great Barrington declaration calling for focused protection, explains about lockdowns:
“It’s a unique experiment, and it’s a terrible experiment. I’m amazed — as are many of my colleagues — at the total focus on this disease. In a short time, we are throwing all the principles of public health out the window. Most countries in Europe had a pandemic-preparedness plan which did not recommend lockdowns, but instead proposed a risk-based strategy to protect those at high risk, which is actually the same as the focused protection we put forward in the Great Barrington Declaration. What we are proposing is, therefore, nothing revolutionary.” [Emphasis added.]
In other words, in the inverted reality in which we now live, the group of scientists I have ironically chosen to call ‘the Radicals’ are advocating for nothing more than the agreed scientific approach while an unscientific and reckless blunt-instrument approach supported by another group of scientists is being pushed in decision-making circles as the only option available.
The illusory quest for total safety
How have we come to the point where an influential body of science academics representing the country’s most prestigious institutes of research and learning are not arguing rationally? John Lee, former professor of pathology and NHS consultant pathologist, lucidly outlines a principle that would explain the proximate cause of this departure from sound reasoning: all bureaucracies, and in particular those dominated by scientists, have been infiltrated by what is known as the precautionary principle. At its core, it states that if a threat to human health is identified, precautionary measures must be taken, even in the absence of scientific knowledge about the origin and scope of the threat, and crucially, without assessment of the harm caused by the recommended interventions. Which is of course unscientific. This goes a long way to explaining the paradox of how a group of highly trained scientists have sought to defend the continuation of crippling restrictions with a letter written in intelligent-sounding gibberish masquerading as science about the elusiveness of facts, the unreliability of research and with exhortations to rely on the murky intangibles of “multiple perspectives” and “interdisciplinary debate”.
In his article, John Lee succinctly summarises the illogical step inherent in the approach:
“So, given that one thing is harmful, therefore something else [the intervention] that is uncertain should be implemented.” [bold emphasis added]
In the case of the Suppressers group led by Greenhalgh, doubts about the usefulness of an uncertain and unproven intervention are then dismissed with: an appeal to fear (Long Covid); meaningless jargon (“complex system dynamics” of the virus); the unproven futility of adopting an efficient and scientific targeted approach based on known risk profiles, and; perhaps most shockingly for a group of scientists, claims about the elusiveness and unreliability of facts and research.
The manure pile of bureaucracy
But if the precautionary principle is a proximate cause of unscientific scientists, what is the manure in which this weed has flourished? Ultimately, I believe it is bureaucratism itself. David Graeber provides insight into the meteoric rise over the last forty years of a class within the burgeoning service sector — administrators. The decline of factory jobs and of traditional working-class occupations that began in the 1980s was swiftly followed by the rise of the professional-managerial class within the services sector. What happened is usually framed as a shift from industrial and manufacturing to “service” work, but, as Graeber points out, this is deceptive since “service” is typically defined too broadly to reveal what’s really going on. In fact, the percentage of the population engaged in what we would regard as productive services — driving cabs or trimming hair, as opposed to the creation of administrative forms and rules — has changed little since Victorian times.
What is not properly acknowledged but hardly contestable is the spectacular growth of clerical, administrative, and supervisory work running concurrently with a growth in “care work”: medical, educational, maintenance and social care. And productivity in the care work sector has actually decreased across the developed world largely due to the weight of bureaucratisation imposed by the burgeoning number of administrators. The effect of this is not only to squeeze the wages of productive care workers but to also make their jobs more difficult since the job of administration and bureaucracy is to perpetuate itself with endless rules, meetings, paperwork, excel spreadsheets and motivational seminars.
But risk management is where the managerial administrator really comes into their own in this story. The reality is that if you are a middle or senior manager in today’s UK service sector, risk management is a major component of your work. And diligent risk management, if it ever existed, has been usurped by the lazy application of the better-safe-than-sorry ideology underlying the precautionary principle. Under this tyranny of the assumption of Murphy’s Law, the newly promoted risk manager quickly learns to over-egg the probability and impact of risks under the better-safe-than-sorry culture driven from the top by board and other senior committees whose biggest fear is waking up to newspaper headlines about a failure to manage a known unknown.
This disproportionate response to risks is underplayed or ignored partly because of the vested interest that middle and senior managers have in maintaining an ever-increasing undergrowth of complex and unintelligible risk registers. Add to that: a ‘willingness’ by the end customer (often, but not always, the taxpayer) to continue paying, and; a zero risk tolerance by senior boards, et voila! The recipe for overcooked risk management. And it would be considered downright reckless to ask for less risk management. But what we should be asking for is better risk management — a proportionate system that automatically forces a balancing of the cost-benefit equation. Failing to count the costs created by over-managed risks is bad management. And bad management by a government on the epic scale we’re seeing now is catastrophic.
Comfort in captivity while bureaucracy perpetuates itself
A strange phenomenon of lockdown is emerging within the professional-managerial class. While many now recognise that the government’s current response in the form of endless conflicting restrictions no longer makes any sense, the IT revolution has afforded them the opportunity to continue filling in forms and analysing data under much improved working conditions at home. They are enjoying their newfound captivity and there is little incentive for many to return to the offices that have come to symbolise the meaninglessness of their working existences. Here’s how the insane circle of bureaucracy completes itself: the response to Covid will have perversely created more work for them as priorities got reshuffled, projects delayed and countless Covid committees set up to monitor the new threats created by the hysterical response to Covid, driven in turn by bureaucracy’s very own precautionary principle. A key lesson to take from this is that the main function of bureaucracy is not to create order but to perpetuate itself. It feeds on itself in order to grow, which makes it devilishly hard to control.
You are what you do repeatedly
The failure of the precautionary principle is not just about what the professional managerial does. It’s also about who they have become. Perhaps the administrators have also lost their ability to empathise with people who are now taking the brunt of the unfolding catastrophe. Graeber sums up the robotic lens through which the professional-managerial class view the world:
“The core value of the professional-managerials might best be described as proceduralism. The rules and regulations, flow charts, quality reviews, audits and PowerPoints that form the main substance of their working life inevitably colour their view of politics or even morality. These are people who tend to genuinely believe in the rules. They may well be the only significant stratum of the population who do so. If it is possible to generalize about class sensibilities, one might say that members of this class see society less as a web of human relationships, of love, hate, or enthusiasm, than, precisely, as a set of rules and institutional procedures.”
If you’re getting paid for doing your job from home, you are not likely to be working in a productive capacity. You are an administrator. Don’t let your high-sounding title or your complex “data analysis” skills delude you into thinking otherwise. The hysterically disproportionate response to Covid has decimated the jobs and lives of people doing real work in the productive sectors, and the government has signalled the indefinite postponement of sanity with its rejection of the Great Barrington Declaration. Bureaucracy’s uncanny knack of feeding off itself is one part of its survival strategy. The other part of its survival strategy rests on its ability to leech off the productive sector. But professional-managerials may not have fully grasped that the goose that lays the egg is being strangled. And not slowly either.
When the many histories of Covid are eventually written, they will portray an event that was not just about a pandemic with facts, statistics and political scalps to pore over. It will be examined for Meaning with a capital ‘M’. No doubt there will be several Meanings and one of them may turn out to be just how far developed economies had been infected by another virus — necrotising bureaucratism — and the damage it did to rationality and truth.
[i] There appears to be a duplication of the name Neil Pearce, which appears twice with exactly the same details.